1. Technical Field
The present invention relates in general to maintaining functionality during component failures. More particularly, the present invention relates to providing component access alternatives to an application when one of the applications encounters an unavailable component.
2. Description of the Related Art
Computer applications are becoming increasingly complex. In the process of becoming more complex, computer applications are also becoming more dependent upon outside components, such as databases and other applications. During a large application's operation, the application may launch other applications and access many databases. In a distributed computer system, an application may launch these components on servers that may be located in a different office complex.
A challenge found, however, is that components that an application depends may become unavailable. For example, an application may attempt to access a database and discover that the database is not responding possibly due to a database interface failure. When an application encounters an unavailable component, the application attempts to recover from the encounter, which typically involves attempting to access the same component a second time. If the application is unsuccessful, the application tends to take an “all or nothing” approach. Either the application completely restarts, or, if the failure is severe, an entire server or set of servers on which the application executes is restarted. In today's business environment where more and more businesses depend upon continuous availability of computer application systems, this is an invasive and time-consuming approach to managing application availability.
Another challenge is developing as systems evolve in support of the extremely dynamic nature of today's business environment. In order to fit this need, applications are becoming less aware of the computer infrastructure on which they run. Technologies such as Virtualization, Automated Provisioning of new servers in real time, and automated business process orchestration make it more difficult to develop component failure contingency plans in advance without a “flexible manager” function to address real outage situations as they arise.
What is needed, therefore, is a system and method for an application to continue operation when the application encounters an unavailable component by offering the application an alternative action to perform.